The Study of Variable Block Size DP-DPCM Image Coding Algorithm for Medical Image Compression
賴健文、吳家琪 ; 鍾翼能
E-mail: [email protected]
ABSTRACT
This thesis is research into improvement of compression for medical image data. The Double Predictor Differential Pulse Code Modulation(DP-DPCM) algorithm used in general images was modified to increase the image transmission speed and signal-to-noise ratio(SNR).The Quadtree segmentation algorithm was adopted in dividing medical images into regions having widely differing perceptual importance. The detail regions of a given image were segmented into blocks with smaller block size, and the background regions of the image were assigned larger block size for each block. Using this rule, we preprocess the image compression as image segmentation. After the pre-processing of Quadtree segmentation, the differential values between the nearby pixels of each much image block are reduced. Therefore, a decrease in the distribution range of prediction error was obtained as well as reduction in the bit rate and quantization levels. In this study, we consider the effect is error free and noiseless in the transmission channel. The conventional DPCM image coding system is easily affect by a larger fed-back quantization error while being transmitted over the channel. The DP-DPCM system can reduce the effect of the fed-back quantization error and does not increase the system complexity. With medical images, however, we are concerned with the diagnostic information and compression ratio. We selected one quadtree block and a high bit rate for simulation. The results are available as a reconstruction image that is acceptable at the low bit rate. Another simulation for the selected quadtree block is using a variable bit rate, where we set the lowest bit rate (1bpp) to saving more bits and transmission time. For medical image compression and reconstruction quality in this system is more flexible in clinical situations. The system performance results of the proposed variable block size DPDPCM image encoder/decoder system are closer to the general image and are better than conventional DPCM.
Keywords : Double Predictor Differential Pulse Code Modulation (DP-DPCM) ; Quadtree segmentation ; bit rate ; background ; detail ; signal-to-noise ratio (SNR)
Table of Contents
封面內頁 簽名頁 博碩士論文授權書．．．．．．．．．．．．．．．．．iii 中文摘要．．．．．．．．．．．．．．．
．．．．．．v 英文摘要．．．．．．．．．．．．．．．．．．．．．vi 誌謝．．．．．．．．．．．．．．．．．．
．．．．．vii 目錄．．．．．．．．．．．．．．．．．．．．．．．viii 圖目錄．．．．．．．．．．．．．．．．．
．．．．．xi 表目錄．．．．．．．．．．．．．．．．．．．．．．xiii 第一章 緒論 1.1 前言．．．．．．．．．．．
．．．．．．．．．．1 1.2 研究動機．．．．．．．．．．．．．．．．．．．2 1.3 研究目的．．．．．．．．．．．
．．．．．．．．2 1.4 論文架構．．．．．．．．．．．．．．．．．．．4 第二章 影像分割與壓縮 2.1 影像分割．．
．．．．．．．．．．．．．．．．．5 2.1.1 不連續性的偵測．．．．．．．．．．．．．．．5 2.1.2 區域導向分割法．
．．．．．．．．．．．．．．6 2.1.3 區域生長法．．．．．．．．．．．．．．．．．7 2.1.4 區域的分裂與合併．．．
．．．．．．．．．．．7 2.2 影像分割的目的．．．．．．．．．．．．．．．．9 2.3 影像壓縮系統的分類．．．．．
．．．．．．．．．10 2.4 壓縮系統的效能評定．．．．．．．．．．．．．．12 2.4.1 信號的客觀量測．．．．．．．．
．．．．．．．13 2.4.2 位元率與複雜度．．．．．．．．．．．．．．．14 2.4.3 壓縮效率的表示法．．．．．．．．．
．．．．．15 第三章 DPCM系統 3.1 PCM通信系統．．．．．．．．．．．．．．．．． 17 3.2 DPCM通信系統．．．
．．．．．．．．．．．．．．17 3.3 線性預測器最佳化．．．．．．．．．．．．．．．20 3.3.1 線性預測器的階數．．
．．．．．．．．．．．．20 3.3.2 線性預測器的最佳化．．．．．．．．．．．．．21 3.3.3 二維一階線性預測器最佳化
．．．．．．．．．．22 第四章 雙預測器DPCM系統 4.1 重建信號預測誤差的發生．．．．．．．．．．．．26 4.2 雙 預測器DPCM的架構．．．．．．．．．．．．．．27 4.3 第二預測器預測係數的推導．．．．．．．．．．．30 4.3.1 誤 差序列e(n)共變異數ρe的推導．．．．．．． 30 4.3.2 誤差共變異數 的推導．．．．．．．．．．．． 31 4.4 DP-DPCM 第二預測器預測係數最佳化．．．．．．． 32 4.5 預測器的預測區間．．．．．．．．．．．．．．．34 第五章 四分樹 分割法與雙預測器DPCM整合系統 5.1 可變區塊四分樹影像分割法．．．．．．．．．．．37 5.2 四分樹影像分割運算法
．．．．．．．．．．．．．39 5.3 可變區塊DP-DPCM編碼系統．．．．．．．．．．． 40 5.3.1 VBDP-DPCM的演算流 程及效能．．．．．．．．． 40 5.3.2 醫學影像四分樹分割的結果．．．．．．．．．．42 第六章 影像壓縮方式的模擬 結果與分析 6.1 模擬程式之處理流程．．．．．．．．．．．．．．44 6.2 預測器的種類及四分樹分割法的臨界值Θ
對SNR的影響45 6.3 壓縮位元率的影響 ．．．．．．．．．．．．．． 55 6.4 平均位元率與重建品質．．．．．．．．．
．．．．62 第七章 結論與展望．．．．．．．．．．．．．．．．81 7.1 結論．．．．．．．．．．．．．．．．．．
．．．81 7.2 未來展望．．．．．．．．．．．．．．．．．．．83 參考文獻．．．．．．．．．．．．．．．．．．．
．．84
REFERENCES
參考文獻 [1] H.-B. Chen ,R.-J,Liu and J.-C. Wu, “Advanced Double Predictor Differential Pulse Code Modulation Image Transmission System
”,1999 [2] J.-C. Wu,H.-B. Chen and R.-J,Liu,“Variable Block—size Double Predictor DPCM Image Coding”,SPIE,vol. 4067 , pp838-847,2000.
[3] D.G. Daut,D. Zhao and J.-C. Wu,“Double Predictor Differential Pulse Code Modulation Algorithm for Image Data Compression
”,Opt.Eng.,vol.32,no.7,pp.1514-1523, July 1993.
[4] J.Vaisey, and A. Gersho, “Image Compression with Variable Block size segmentation”, IEEE Trans.on Signal Processing, vol. 40, no 8, pp2040-2060, Aug 1992 [5] Rafael C. Gonzalez and Richard E. Woods, Digital Image Processing,Addison-Wesley Publishing Company,1992 [6]
C.S.Won, “ Variable block size segmentation for image compression using stochastic models ,”Image Processing 1996 Proceedings, International Conference on Volume : 3,pp.975-978,1996 [7] C.T. Chen,“Adaptive Transform Coding via Quadtree-Based Variable Blocksize DCT,
”Proc.ICASSP,pp.1854-1857,May 1989.
[8] R.Distasi, M.Nappi and S. Vitulano,“Image Compression by B-Tree Triangular Coding. ”IEEE Trans.on commu.,vol.45,no 9,pp.1095-1100,Sep,1997.
[9] C.Y.Teng and D.L.Neuhoff,“A new quadtree predictive image coder, ”Image Processing 1995.Proceedings. ,International Conference on vol.2,pp.73-76,1995.
[10] 劉晟志,黃煥超 編譯,張隆紋 校閱, “數位影像處理”,儒林圖書,1988 [11] 戴顯權,陳文儉, “資訊與教育雜誌”,Jun,2001 [12] 陳培殷,
“資料壓縮概論”,滄海書局,2001 [13] M.J. Sabin,“DPCM Coding of Spectral Amplitudes without Positive Slope Overload ”, IEEE Trans.on Signal Processing, vol. 39,no 3, pp.756-759, Mar,1991 [14] K.Sayood,“Introduction to Data Compression”,CA: Morgan Kaufmann,2000 [15]
N.Jayant, “Signal compression:Technology targets and research directions,”IEEE on Selected Areas in Communications,vol.10,pp796-818,1992 [16] 吳樂南編著,徐孟俠審,“資料壓縮的原理與應用”,儒林圖書,1996 [17] F.G,Stremler原著,周俊賢編譯,“通訊系統導論”, 全華科技圖 書股份有限公司,pp380-391,1991.12二版 [18] 張傳生著,“數位通訊原理”,儒林圖書,1992 [19] 戴顯權,“資料壓縮”,松崗書局,1999 [20]
N.S. Jayant and P. Noll, “Digital Coding of Waveform” ,Englewood Cliffs,NJ,Prentice-Hall,1984 [21] M.C. Rost and K.Sayood, “An Edge Preserving Differential Image Coding Scheme”, IEEE Trans.on Image Processing, vol. 1,no 2, pp.250-256, 1992 [22] D.G. Daut and J.-C. Wu,
“Adaptive Cosine Transform Image Coding with Variable Block Size and Constant Block Distortion”, in the 1996 Symposium on Visual Communications and Image Processing, SPIE vol.2727,pp. 1104-1115,March.1996 [23] J.-C. Wu and D.G. Daut, “Adaptive Nonstationary DPCM Image Coding with Variable Blocksize”, in The 1997 Symposium on Visual Communications and Image Processing, SPIE vol.3024,pp.
447-458,Feb.1997
